VM with HA messes up WiFi

Hi there!

I’ve quite the interessant problem/puzzle to solve. Running a vm with HAOS and home assistant, my wifi devices(Tuya, esp and other looks like 2.4Ghz devices) sometimes are completely unavailable. Only thing that works is a reboot of the vm and all the unavailable WiFi devices come up instantly. Rebooting just home assistant doesn’t work and rebooting the switch and router neither. Happens sometimes after 6 hours and sometimes after three days.

Any thoughts?

my setup:
Minisforum GK50
Unraid 6.12.11
VM standard network settings
Tplink deco XE75PRO mesh network (with about 45 connections, hardwired and WiFi)
Home assistant 2024.7.3
HAOS 12.4

Thanks!

Hi, have you checked if you have no IP-address conflict?

Why are you running a VM?

I have set the vm a fixed ip and also reserved that in my deco.

More versatile and low power usage/high performance.

Huh?

Why are you using a VM? Why not bare metal HAOS?

He literally just told you why.

2 Likes

You gave your HA VM a fixed IP + reserved that in your AP?!?
Isn’t that causing an IP conflict?

And do I understand it correctly: your HA VM is the cause that your WiFi devices become unavailable?
So you can’t even ping them from your pc?

Features are not a “why”.

My mistake, just standard network settings in VM, no fixed ip.

I cannot ping the devices. Also in the Tuya app the devices become unavailable. It took me a while (and quite some frustration :grimacing:) to narrow it down to the vm with HA. When I reboot that, everything becomes instantly available.

Which is why I asked, why VM and not bare metal. HAOS on bare metal has none of these silly IP problems.

Have you also retried restarting just the integration? And are you using local tuya integration, or cloud based? And what about the esps, are they just showing unavailable in esphome, or actually not connected to HA? It sounds like you could be having mdns or cloud service trouble, not “ip” or wifi problems.

Why are we virtualizing hardware? It’s a cost and energy effective solution. And I don’t want to have my utility cabinet clogging up with raspberries and nucs :grimacing:

1 Like

That’s just the thing, it doesn’t matter if I restart any integration or Home Assistant itself. I have to restart the whole virtual machine and then instantaneously everything works again. I have also have problems with another WiFi device (my WiFi stick for my solar inverter) which isn’t connected to home assistant directly.

Edit: I use cloud based tuya and the standard tuya integration. Esphome, I’ll check next times it goes offline.

1 Like

That actually helps. It tells us the VM is the contributing factor not haos or the integration.

And what about other network connected devices and cloud services? Because in reality, nothing connects to HA using wifi. So pointing at wifi (if it really is wifi) is putting the cause outside HA.

1 Like

I’m sorry, what? Your anti-virtualization crusade makes no sense.

1 Like

There was someone else within the last few weeks with a complaint about losing connectivity after haos 12.4 but I can’t find it right now and unfortunately it had very little details.

Op tell us about your IP network.

What IPV4 address space are you using. Is ipv6 turned on? What ip is assigned to the router. The settings for the VM hypervisor host nice and the IPV4 settings for HA inside the VM.

Router ip: 192.168.1.1
Subnet mask: 255.255.255.0
DHCP range: 192.168.1.50-192.168.1.250
Home assistant vm: 192.168.1.93

Unraid vm setting: network source: br0, network model: virtio-net
No ipv6 enabled on unraid

Home assistant

IPv4

IP-adres: 192.168.1.93/24

Gateway: 192.168.1.1

Methode: auto

Naamservers: 195.121.1.34, 195.121.1.66

IPv6

IP-adres: fe80::e9be:98fb:9989:d36/64

Gateway: fe80::1e61:b4ff:feac:6141

Methode: auto

What happens if you ping 192.168.1.93 and turn off the vm while you are pinging? Since you are using an IP in the dhcp range it could already be in use by something else.

1 Like